[Zurück]

G. Mandlburger:
"LiDAR Data Management from a trans-national perspective";
Vortrag: ESF Exploratory Workshop, Heidelberg (eingeladen); 09.09.2011 - 10.09.2011.

Within the last decade, the availability of countrywide Airborne Laser Scanning
(als) datasets featuring high resolution (>1 pt/m2) and height accuracy (<15 cm)
has steadily increased. In Europe, many countries are currently working on finishing
their national als campaigns (e.g. at, de) whereas others are already updating
their initial datasets with higher densities (e.g., nl). The point cloud as primary
deliverable of als is the basis for many subsequent models (e.g., dtm/dsm) and
applications (e.g., city modeling, forestry, infrastructure, hydrology, disaster man -
age ment, ecology...) and, therefore, an efficient seamless management is of
crucial importance. This contribution reviews the aptitude of different geo-database
for the administration of trans-regional als point clouds and derivatives.
Commercially available products (Oracle Spatial) are discussed as well as open
source solutions (Postgres/Postgis) and scientific approaches (Topdm). High performance
spatial indexing methods (Kd-, R*-, Quad-, Oct-Trees) are a preconditions
for an efficient data management but, beyond that, organizational matters (data
hierarchies, metadata, spatial and temporal reference systems) are crucial. For a
European Laser Data Spatial Data Infrastructure initiative, especially the smooth and reproducible 3d-transformation from the European Reference System (etrs
89/evrs) to the respective national geodetic systems (and back again) is of
high importance as current als data has to be combined with (historic) national
geo-data. This requires compliance to approved standards (e.g. ogc Coordinate
Trans formation Service) and their application, but, however, the standards lag
behind the practice as als data in sub-dm accuracy often require model oriented
transformation approaches (NTv2 grid shifts, height systems...), especially in a
trans-national context.